1. Field of the Invention
The present invention relates to moving picture coding and decoding techniques, and more particularly, to moving picture coding and decoding techniques using motion compensation prediction.
2. Description of the Related Art
As a representative compression coding scheme of a moving picture, there is an MPEG-4 AVC/H.264 standard. In MPEG-4 AVC/H.264, motion compensation of partitioning a picture into a plurality of rectangular blocks and predicting a motion from a reference picture using an already coded or decoded picture as a reference picture is used. A technique of predicting a motion by motion compensation is referred to as inter prediction. In the inter prediction of MPEG-4 AVC/H.264, a plurality of pictures can be used as a reference picture, and an optimal reference picture is selected for each block from among the plurality of reference pictures, and then the motion compensation is performed. To this end, a reference index is allocated to each reference picture, and the reference picture is specified by the reference index. In a B picture, a maximum of two reference pictures are selected from coded or decoded reference pictures and used for the inter prediction. Prediction using the two reference pictures is classified into L0 prediction (list 0 prediction) mainly used as forward prediction and L1 prediction (list 1 prediction) mainly used as backward prediction.
Further, bi-prediction simultaneously using two inter predictions, that is, the L0 prediction and the L1 prediction is defined as well. In the bi-prediction, a final inter prediction picture signal is constructed by performing predictions of both directions, multiplying signals that have been subjected to the inter predictions of the L0 prediction and the L1 prediction by a weighting coefficient, adding an offset value, and causing the resultant signals to be superimposed on each other. For the weighting coefficient and the offset value used in the weighting prediction, a representative value is set and coded in units of pictures for each reference picture of each list. Examples of coding information related to the inter prediction include a prediction mode representing the L0 prediction, the L1 prediction, or the bi-prediction for each block, a reference index specifying a reference picture for each reference list of each block, and a motion vector representing a moving direction and a moving amount of a block, and this coding information is coded or decoded.
In a moving picture coding scheme in which motion compensation is performed, in order to reduce a coding amount of a motion vector constructed in each block, a prediction process is performed on a motion vector. In MPEG-4 AVC/H.264, a motion vector predictor is derived by performing prediction based on a neighboring block using the fact that a motion vector of a coding target has a strong correlation with a motion vector of a neighboring block, a motion vector difference that is a difference between the motion vector of the coding target and the motion vector predictor is derived, and the motion vector difference is coded, whereby a coding amount is reduced.
Specifically, as illustrated in FIG. 36A, a median value is derived based on motion vectors of neighboring blocks A, B, and C and set as a motion vector predictor, a coding amount of a motion vector is reduced by obtaining a difference between the motion vector and the motion vector predictor. Here, when the coding target block is different in size and shape from the neighboring block as illustrated in FIG. 36B, an uppermost block is decided as a prediction block when a plurality of neighboring blocks are on the left side, and a leftmost block is decided as a prediction block when a plurality of neighboring blocks are above the coding target block, and prediction is performed based on the motion vector of the decided prediction block.
Patent Literature 1: JP 2011-147172 A
However, in the method of the related art, since only one vector predictor is obtained, there are cases in which, depending on a picture, a prediction accuracy of a motion vector predictor is low, and the coding efficiency is not improved.
In this regard, the present inventors have recognized that it is necessary to further compress the coding information and reduce the entire coding amount in the moving picture coding scheme using the motion compensation prediction.